INDEX

Explanations

mentions of specific people or topics in a text

oai_token-act-pair · gpt-3.5-turbo

terms related to safety and risk

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ILCS

-0.65

 COURT

-0.50

soever

-0.47

 partName

-0.47

 Ballistic

-0.45

 ORDER

-0.45

 Hubble

-0.44

iquid

-0.44

)--

-0.44

pires

-0.44

POSITIVE LOGITS

perty

0.52

viously

0.51

rats

0.50

etz

0.49

Semitism

0.48

 feasibility

0.48

addon

0.48

 plight

0.48

orth

0.47

 hotline

0.47

Activations Density 1.381%

mentions of specific people or topics in a text

terms related to safety and risk

No Comments

No Known Activations